Open-ended rectangular waveguide probes have demonstrated utility for many applications in the field of microwave and millimeter-wave nondestructive testing and evaluation. Such testing and evaluation includes three-dimensional imaging, crack detection, and material characterization. In many instances, using an open-ended rectangular waveguide for material characterization is preferred over other techniques such as loaded transmission line or cavity techniques that require the material being tested to be cut and shaped to fit inside the transmission line or cavity.
FIG. 1 illustrates a probe 100 for use in determining electrical and magnetic (complex dielectric and permeability) properties of materials. The probe 100 comprises an open-ended waveguide 102 having a finite flange 106. In operation, a signal source provides a microwave signal to the waveguide, which in turn transmits microwave electromagnetic energy incident upon an object to be tested. The microwave electromagnetic energy penetrates the object placed in front of the open end of the waveguide based on the object's dielectric properties. Cracks and other surface variations affect the dielectric properties and are thus detectable by the probe.
When used for material characterization, conventional open-ended rectangular waveguide probes, such as the probe 100, require robust full-wave electromagnetic models along with optimization algorithms to achieve acceptable accuracies. Such electromagnetic models provide the reflection coefficient at the aperture of the waveguide radiating into a dielectric structure. But any small error in modeling and/or measurement of the reflection coefficient may lead to unacceptable errors in estimating the dielectric constant or thickness when measuring the dielectric properties and thicknesses of thin and low loss materials, especially when they are embedded within a structure containing thicker and lossy dielectrics.
Even if an electromagnetic model accounts for generated higher-order modes and is generally accurate, the finite flange 106 contributes to the majority of the error in estimating the dielectric constant and/or thickness of a layer or layers within a stratified dielectric composite structure. This is due to the fact that the model assumes an infinite flange, while measurements are commonly conducted using waveguides with standard finite flanges. Moreover, through extensive measurements it has been shown that the adverse effect of the finite flange 106 is more significant for estimating the dielectric constant or thickness, especially for thin and low permittivity and low loss materials. This adverse effect has further significance for conductor-backed composite structures.
Consequently, when performing measurements using a conventional open-ended rectangular waveguide, a very large flange is sometimes used. Otherwise, testing must be limited to primarily lossy materials or else the errors introduced by flange 102 cause poor measurements. In other words, accurate measurements of low loss material characteristics cannot be taken by a conventional open-ended rectangular waveguide such as probe 100. Although a lossy dielectric sheet may be used as a coupling medium to reduce the flange effect, the extra attenuation introduced by this additional lossy sheet reduces the measurement sensitivity.